What is the name for the part of our ear that vibrates when it gets hit by a soundwave?”
I dream about this question now, and the long pause which follows, as all five students avoid the teacher’s gaze. I’ve seen 89 trainee primary teachers ask it (twice each!). I’ve watched them respond to their students’ silence: some hint, some nominate, some keep talking, hoping a student will eventually raise their hand. No one ever does.
It was an experiment (described here). We invited trainees to ask five Year 4 pupils revision questions about sound. We offered guidance encouraging them to cold call, wait three seconds before asking a student to respond, correct wrong answers, and offer a hint if students couldn’t respond.
The experiment controlled possible variables. We gave the trainees the questions and the answers. Trainees were teaching avatar pupils (shown below), voiced by a simulation specialist. The specialist had a script detailing when and how to respond to different prompts: when to say “I don’t know, sorry,” for example, and when to relent and offer an answer.
Controlling the variables – setting identical questions and responses – and repeating this 178 times, allows you to identify patterns in how teachers ask questions in a way that would be almost impossible through classroom observations. Many trainees pursued these retrieval strategies effectively, particularly after seeing a video model (a finding described here). But I rapidly noticed a handful of common mistakes: tempting but unhelpful ways to ask questions. This post describes the five most prevalent ways not to ask questions for retrieval – and what to do instead.
Mistake 1: Asking for hands up
Most trainees invited and relied on hands up:
- “Who can remember…?”
- “Hands up if…?”
- “Good to see some more hands here.”
What’s going on here? Putting hands up is a signature feature of classroom teaching. Trainees do it almost automatically. But a signature feature may not be desirable. In particular, when we’re asking students to retrieve existing learning, the goal is not “Who can remember…?” but “Can we all think back…?” On one question, our simulation specialist’s script said not to raise any of the students’ hands. Some trainees seemed genuinely unsettled as a result. More broadly, relying on hands up means we rely on the most confident or interested students responding. They learn more, while less confident and less motivated students are left out. If they know they won’t be called on, they don’t need to think about the response. We create a two-tier classroom: those who are participating directly and those who aren’t. Some trainees would not nominate a student unless someone raised a hand. When no student raised a hand, some simply kept talking, for two or three minutes even, encouraging, hinting, and eventually giving the right answer, without students saying a word.
A better option… is to nominate a student to respond:
- “I’m going to call on someone to tell me…”
- “I’ll be cold calling…”
- “Hands down, I’m going to choose…”
This gets every student thinking when a question is asked, and involves all students in answering and so participating in the lesson at some point. In doing so, it boosts students’ confidence and participation.
Mistake 2: Mostly asking the boys
- “What do you think Will?”
- “Let’s call on Will.”
- “Will, can you tell us?
What’s going on? If I hear Will called upon one more time, I’ll scream. Was it because he was kind-of in the middle? Was it because he was bigger, looked older, or was white? Whatever the reason, trainees seemed disproportionately keen to hear from him.
In particular, trainees seemed to call on him after another student had got the answer wrong. However we choose to explain the underlying dynamics there is decades of evidence that boys, white students and high attainers talk more in the classroom (Howe and Abedin, 2013). If we’re not actively working against that to make classroom participation fairer, we’re accepting it instead.
A better option… is to select who participates. We could:
- Preselect students to ask questions – I’m not wild about this, since I tend to come up with most questions on the spot. But many teachers I deeply respect are adamant you should be planning your questions, so maybe you should. If you plan questions, you can plan respondents.
- Many teachers I deeply respect are dead against this, but I still believe lollipop sticks, or some kind of random selection tool, are an invaluable technique for fair participation, and way to communicate the classroom you’re trying to build.
- Alternatively, you could use a register to track who has talked as the lesson has gone on, and use this to target questions (Will has spoken five times, Mina zero, so my next question goes to Mina).
Mistake 3: Bouncing the question to another student (too early)
Once teachers had picked someone to call on, if a student got an answer wrong, or said they didn’t know, most trainees immediately called on another student to answer:
- “Can anybody help Carlos?”
- “Carlos would you like to phone a friend?”
- “OK, let’s ask Mina instead.”
What’s going on? The impulse here seems to be an urge to get to the right answer. This may be combined with the teacher wishing to avoid embarrassing the student called upon. But the result is that the student called upon is rarely encouraged to have a go, or to improve their initial answer. Longer term, this may accustom students to (a) giving ‘first draft’ answers, and (b) opting out if they don’t want to try. It’s particularly unhelpful if your goal is retrieval: all students have encountered the right answer – we need to push them to recall it. Bouncing to another student – the famous “Pose, pause, pounce, bounce” can be useful in sparking discussion, but only after the first student has offered a substantive answer to discuss.
A better option… would be to encourage students, push them, or offer them a hint:
- “What could it be Carlos?
- “What can you remember about this question?”
- “The part of the ear that vibrates shares its name with a musical instrument.”
Mistake 4: Don’t round up a wrong answer to a right one
When students gave a wrong answer, we often heard:
- “That’s nearly right, we just need to also mention…”
- “Good answer, can anyone add to that…”
- (After calling on another student) “So if we stick that together with what you said, we’ve got the right answer.”
What’s going on? Teachers were trying to be nice. They succeeded: they communicated to students that they had got the right answer, or as near as makes no difference. But this wasn’t true: we had scripted the answer; it wasn’t right. The risk is that students hear “Good answer,” and assume their answer is indeed good. There’s a tension between protecting students’ egos and getting to the right answer. The trainees I watched seemed much more concerned to protect egos than get to the right answer. But making students feel good about themselves is only part of a teacher’s job. And we can make them feel good about themselves far better if we do so based on students’ genuine achievement.
A better option… is to protect the ego, while being clear about the gaps. And/or to offer praise after students have got the right answer:
- “Thanks Mina, that’s helpful, can you add to that, mentioning…”
- “That’s not quite right, I’m sure you can do it, have another think…?”
- “That’s not quite right, can you have another go… Good, answer, well done!”
Doug Lemov has suggested that the reward for a right answer should be another question. Perhaps this is true for a wrong answer too – at least if we know the students have recently studied this topic.
Mistake 5: Talking too much
Ideally, we wanted teachers to pause for three seconds of silent wait time between the question and nominating a student to answer. Much more often, we heard something like this:
- “What is the name for the part of our ear that vibrates when it gets hit by a soundwave? So I’m looking for the part of the ear – it’s a part of the ear – that vibrates when it gets hit by a soundwave. A part of the ear. What do we think? Does anyone want to have a go? I’ll call on… Jayla.”
What’s going on? Trainees probably thought they were helping by repeating the question. But it felt more like they were filling the silence. A three-second pause for silence, if you’re not used to doing it, seems wildly unnatural (as I found when I tried to make myself do it). As with all these techniques, there’s a time to repeat the question. There can be value in narrating what you see. “Good to see so many hands up,” (if you are using hands up) emphasises a positive social norm. But if the goal is to promote retrieval, to promote individual, effortful thinking, we need to let students actually do that thinking.
A better option… is to say the question clearly and slowly the first time. Pause for three seconds. Then repeat the question if it’s really needed. Then nominate a student to respond. If the student doesn’t understand, you can always say the question again.
Conclusion
I feel uncharitable. The trainees who worked with us had a tough gig: three months into training, five minutes to read about retrieval questions, and two chances to make it work. Many of them did really well. When we showed them a video model of effective questioning, they got much better.
But their struggles offer insight into what new teachers – all teachers, perhaps – struggle with about retrieval questioning. I think these patterns are common enough, interesting enough, and important enough, to report.
These patterns tell us something about teacher learning.
First, their commonality in novice teachers persuades me that what I’ve called here ‘mistakes’ are in fact fairly innate elements of human behaviour when teaching. Perhaps they’re biologically primary: they reflect basic interactional patterns we learn early and untaught. Or perhaps they reflect the ‘apprenticeship of observation’ teachers perform, spending years watching teaching, and learning how teaching works, before becoming a teacher themselves. Perhaps it’s both. Teachers will cleave to this model unless or until they are shown something different – and manage to habituate themselves to do it.
Second, the instinct of most trainees, and maybe most teachers, is to avoid children feeling bad. This innate reaction unifies these slipups: trainees are trying to avoid discomfort and social awkwardness. So they pick students who want to answer. They move on quickly where a student says they’re stuck. They round up answers. And they keep talking until they think students will have the answer.
A teacher who makes their children feel bad will struggle to teach them anything. But the job of a teacher is to push students beyond what they think they can do. As Deborah Ball and Francesca Forzani have argued, to teach effectively we may have to suppress our instinctive reactions, and our preferences. This may mean teaching books we dislike, welcoming parents we find tricky – and probing where students are uncertain (2009, p.499). The goal was retrieval: trainees would have done better if they had pushed harder for retrieval.
This is hard. But pushing students to do things they may not feel ready or able to do unlocks the door to what Colin Burrow recently described as the main pleasure of teaching: “gently persuading someone else to recognise in themselves what you can see in them.”
If you enjoyed this, you might appreciate…
- I tried, and struggled, to increase wait time in my teaching, and described the process and results here.
- I defended the practice of using lollipop sticks to call on students here.
- I’ve described the wider experiment, and what we learned about using models to help the trainees teach better, here.
I think there are even better methods when the question is a deeper one. My book Teaching Today 5th Ed OUP describes them but there is a training activity on my website I’ve used hundreds of times and it works every time to assess different questioning methods with important criteria:
https://geoffpetty.com/training-materials/questioning/